Self-tuning vibration absorber based on model reference adaptive control

Abstract

The performance of a tuned vibration absorber (TVA) is highly sensitive to its structural parameters, and even slight mistuning can significantly degrade its vibration suppression performance. However, structural parameter uncertainties caused by manufacturing tolerances, along with time-varying features such as parameter drift or abrupt changes due to aging or environmental variations, often result in TVA mistuning and degraded performance. Therefore, developing precise tuning strategies for TVAs with uncertain and time-varying structural parameters is of great importance, and further research is urgently needed in this area. To address this challenge, this paper proposes a self-tuning vibration absorber based on model reference adaptive control (MR-STVA), which possesses the ability to compensate for structural parameter uncertainty, drift, and abrupt changes − a capability we define as structural parameter adaptive compensation. First, an adaptive control law is developed using Lyapunov stability theory to ensure system stability. Then, a fixed Lyapunov matrix is constructed by leveraging the specific structure of the TVA, allowing MR-STVA to track dynamic reference models and maintain effective suppression of time-varying disturbances. Moreover, a non-uniform gain scaling strategy is proposed to suppress undesired variations in the adaptive control law and reduce energy consumption. A series of experiments are conducted on a custom-built platform, including: (1) self-tuning under structural parameter uncertainty, (2) self-tuning under abrupt stiffness changes, (3) adaptive vibration absorption under time-varying excitations, and (4) convergence performance evaluation. Results verify the effectiveness and practical feasibility of MR-STVA under complex and variable operating conditions.